SMALL-SCALE COGENERATION FOR BUILDING APPLICATIONS PART 2 - OPTIMAL SIZING OF THE CHP PLANT

Jacek Kalina, Janusz Skorek Division of Thermodynamics and Gas Energy, Institute of Thermal Technology, Silesian University of Technology, 44-100 Gliwice, Konarskiego 22, Poland Summary In the paper the problems of optimal sizing of a small-scale CHP plant for building energy supply system are presented and discussed. Calculation procedures are shown together with the results of sample analysis. The factors that influences the final configuration of the system are indicated. Keywords: cogeneration, buildings, gas engines, gas turbines, electricity and heat load variations, optimization, gas boilers. INTRODUCTION In this paper the problem of optimal sizing of a small-scale CHP plant for building energy supply system is being discussed on the example of demonstration project. The project consist in design and installation of CHP plant that would meet the heat and electricity demands of the group of three different buildings. The energy demand analysis has been presented in the part 1 of this paper. The sizing procedure of the plant is based on daily heat and electricity load profiles. Schematic diagram of the heat and power plant is shown in figure 1. It has been assumed that the system is based on the natural gas fired reciprocating engine. It was found from the energy demand analysis that only microturbines or fuel cells can be an alternative solution in this case. However, both turbines and fuel cells are much more expensive than engines, what makes an economic effect of the project worse. In order to meet heating demand of the consumers the plant produces hot water that is subsequently fed into the local heating network of 90/70 OC temperature characteristics (at external temperature -20OC). In case of installation of the CHP module, the electricity is transmitted via common bus to either the consumers or to the external grid operated by the electricity company. In the case of electricity shortages, the bus can be also supplied with power from the grid.

Figure 1. Cogeneration plant (CHP cogeneration module, B gas boiler; VC ventilator cooler, HA heat accumulator, G - generator) The objective function for the plant optimisation procedure is the Net Present Value of the project for specific time of economic life (typically N = 15 years) [1,1,3]:

CFt (1) J 0 max ( 1 + r )t t =1 The objective function is typically constrained by technical parameters of the plant and the economic environment. Therefore the complete analysis consists of two integral parts: technical analysis and economic analysis. Scheme of the optimization algorithm is shown in figure 2. NPV =
N

Figure. 2. Block diagram of plant sizing optimisation procedure Momentary energy balance for the entire CHP plant can be written as follows:

& +Q & & W + N = N ( 1) N + Q (2) P d G D S D str Then particular elements of the equation 1 can be decomposed as follows:

N D = N CHPi +N G + ( 1) N S N Pj
i =1 k =1

nCHP

nP

(3) (4)
+
nK

& = Q D

nCHP i =1

& & & & Q CHPi + QKj Qstr + Q

j =1

nK

&W = P d

nCHP i =1

&W ) & (P d CHPi + ( PWd ) Kj =

j =1 i =1

nK

nCHP

N CHPi
E _ CHPi

& Q Kj

j =1 Ekj

(5)

Ratio of electricity and heat for CHP plant is expressed by the cogeneration index : N (6) i = & CHPi QCHPi The minimum value of the cogeneration index is limited by the maximal possible heat output of the particular gas engine. Particular parameters of the machinery characteristics (E_CHP, ) typically depend on the nominal electric power and momentary load. Figure 3 presents energy characteristics of cogeneration modules with gas engines worked out for the purpose of preliminary machinery sizing procedure. a) b)
45

1,2 1 wskanik skojarzenia 0,8 0,6 0,4 0,2 0

0 1000 2000 3000 4000 moc elektryczna, kW 5000 6000 7000

40 sprawno , %

35

30

25

= 23,485 N0,0691 R2 = 0,7203

= 0.3605 N0.1137 R2 = 0.5821

0 1000 2000 3000 4000 5000 6000 7000

20

moc elektryczna, kW

Figure 3. Efficiency (a) and cogeneration index (b) of gas engine based CHP modules in the electric power range of 50 6000 kW (note: exhaust gases cooled down to 120 OC) In order to estimate heat output and fuel energy consumption at partial loads the following dimensionless characteristics of the CHP module have been used: E _ CHP (7) 3 2
( E _ CHP ) nom = 0.0025( el ) 0.2431( el ) + 0.587( el ) + 0.6537

3 2 = 0.8147( el ) 1.9848( el ) + 1.7756( el ) + 0.3968 nom

(8)

where:

el =

N el N el _ nom

- dimensionless load, el = 0.2 1.

Efficiency of gas boilers is mainly a function of heat transfer area. Therefore for each nominal heat output of the boiler the same value of efficiency can be

reached. it was assumed that this values is fixed at the level Ek = 0.92 and it does not vary with the nominal heating power of the boiler. In order to estimate the behaviour of the boiler at partial loads the following dimensionless characteristics has been used: Ek (9) 4 3 2

( Ek )nom

= 0.1318 x + 0.1693 x 0.1096 x + 0.1165 x 0.9556

& - dimensionless load. where: x = Q & nom Q

In was found from the heat load duration curve that the maximum heat demand occurs only during a very short period of time over the year. Therefore it is needless to size the heating system for maximum load as it will mostly operate at partial load (with lower efficiency). It is usually possible that the building designer allows shortages of heat in some periods. This allowances must be known at the stage of planning the heating system. The temporary heat shortages & can be included in energy balance of the plant: Q & = Q & (10) Q D The index must be controlled over all period of the simulation of plant operation. If the value of the index or the time of its appearance exceed the set up limits the system must be configured again. Each machine or device can operate only within the defined range of allowable load. It means that the following inequality constraints have to be taken into account: (11) (N CHPi )min N CHPi (N CHPi )nom
& ) (Q
Kj min

& Q & Q Kj Kj

( )

nom

(12)

If the local cogeneration system has to meet simultaneous demands for heat and electricity there can be defined 9 cases of relations between the heat and electricity demands and the production capacities of the plant [1] It was found that the priority of the system operation strongly influences the final economic effect of the project [1]. There are several possible modes of CHP module operation. In this paper the following modes are taken into account: 1) electricity tracking (ET) in this mode the priority is electricity production. The power of the cogeneration module is following the demand of the consumer. There is no transfer of surplus electricity to the external utility grid. Heat demand is balanced by the boiler or heat storage tank. If there is a surplus heat it is dissipated into the atmosphere. 2) heat tracking (HT) in this mode the priority is heat production. The heat output of the cogeneration module is following the demand of the consumer. CHP module usually operates in parallel with boilers. Electricity demand is balanced by the grid.

3) full load operation (FL) cogeneration module is run at full load no matter what the demand of the consumer is. The momentary energy balance of the plant converges by the cooperation with grid, boilers, storage tanks, ventilator coolers and other devices. Almost each cogeneration module equipped with standard automatic control system can run in the above defined modes. Optimization of the plant was done by searching for the best solution within the predefined range of trial variants. Energy balance calculations were performed as an hour by hour simulation of the plant operation [1,2]. The analysis was carried out with using an Excel spreadsheet and Visual Basic macros. Columns of the spreadsheet (matrix) represent particular positions of equations (2) (5), whereas the rows represent time. Once the annual energy balance of the plant is done the economic analysis starts. The economic analysis basis on the cash flow CF calculation for the whole lifetime of the project: N N (13) CF = CF = ( J + S ( K + K + S ) P + L )

t =0

t =0

op

In order to estimate investment costs J0 the typical curves of unitary cost i were used: a) for gas boilers (boiler purchase cost, z/kW): & 0.13 (14) i = 250Q nom
& where Q denotes nominal heat output in kW. Heat output range 50 kW to nom 10000 kW. b) cogeneration module with gas engine (together with ventilator cooler) unitary cost in US$/kW within the electric power range 9 kW do 6000 kW: (15) i = 2594.9( N nom ) 0.2857

All additional costs were estimated either with using a typical cost breakdown or the offers from vendors. It appeared that equipment purchase cost lays down typically in the range of 40% to 60% of the total investment cost J0. In our case real incomes appear only for HT or FL modes of cogeneration module operation. The income results from the sale of the electricity surplus to the grid and it is calculated as follows:
S n = N S d
0

(16)

Total exploitation cost can be expressed as follows: K E = K en + K O & M + K p + K r The most important is the cost of energy:
K en =
R
0

(17) (18)

& K

en

R nCHP nK & Q N CHPi Kj k fk + N G k el ( ) d d = k + i =1 LHVCHP E _ CHPi fCHP LHV = 1 j k Ekj 0

Operating and maintenance costs were calculated with using the typical index of unitary costs for gas engines kO&M = 0.007 do 0.02 US$/kWh (at annual availability in the range of 92 97 %). Therefore the following equation can be used:
K O & M = Eel kO & M = kO & M
R
0 n R

R nCHP
0 i =1

CHPi

d
R
0

(19)

Environmental costs are given by equation:

& & & d = G K env = K env Pi _ k k Pi d + GW kW d & Pi _ CHP + G
i =1 0

[(

) ]

(20)

& Pi _ CHP , G & P _ B are calculated with using emission indices for Typically G particulate machinery type. Emission fees in Poland are regulated by the government: SO2 - 0.38 PLN/kg; CO2 - 0.00020 PLN/kg; CH4 - 0.00020 PLN/kg; NO2 - 0.38 PLN/kg; CO - 0.10 PLN/kg; NMHC - 0.10 PLN/kg; dust - 0,25 PLN/kg (Note: 4.7 z = 1 EURO).
REFERENCE CASE ANALYSIS

In the reference case of calculations it has been assumed that no cogeneration module will be installed on site. The existing coal fired boiler plant will be replaced with gas boiler plant, electricity will be purchased from the utility grid separately for each building. Proposed gas boiler plant would consist of three gas boiler of respectively 350, 350 and 200 kW heat output. Fuel and energy prices, that were used in the analysis, are as follows: Natural gas price depends on an annual consumption, average value is: 0.796 PLN/Nm3. Electricity price: - sport centre (tariff C21): 328.67 PLN/MWh, - school (tariff C11): 336,17 PLN/MWh. Technical and economic results of the project are given in table 1. Table 1. Results of technical and economic analysis for base case project
No. 1 2 3 4 5 6 7 8 9 Quantity Amount of heat produced at boiler plant Heat shortage Total amount of electricity from utility grid Total amount of natural gas burned Average efficiency of the boiler plant Total cost of electricity Total cost of natural gas Investment cost Net Present Value after 15 years Unit GJ/a GJ/a kWh/a Nm3/a % PLN/a PLN/a PLN PLN Value 7272 54 661 637 230 871 90.7 218 600 183 750 466 540 -3 463 600

PLANT WITH COGENERATION MODULE

Table 2 presents the configurations of the cogeneration plant that were defined as trial solutions for optimization procedure. Table 2. Selected variants of the configuration of the plant
Nominal parameter NCHP kW & kW Q CHP & kW Q K1 & kW Q K2 & kW QK 3 1 20 44 350 350 160 2 30 62 350 350 140 3 40 79 350 300 170 4 50 96 350 300 155 Number of variant 5 6 7 60 70 80 112 128 143 350 350 350 300 285 275 140 140 140 8 90 158 350 260 140 9 100 173 330 260 140 10 110 188 300 260 150 11 120 203 300 250 150

It was assumed that there will be only one cogeneration module. The minimal allowable electric load of the module was set in the value of 40% of nominal power. Figures 4 and 5 show the results of technical analysis. In the electricity tracking mode there is no sale of the electricity surplus electricity to the grid, however if the installed electric power of the CHP module is higher, the production is much lower than potentially possible (compare to FL mode). On the other hand in the ET mode the loss of heat form CHP module is lower.
1100 1000 900 800 700 MWh 600 500 400 300 200 100 0 20 30 40 50 60 70 80 90 100 Nominal electric power of the CHP module 110 120
Gross electricity production - ET mode Gross electricity production -HT mode Gross electricity production - FL mode Electricty sold to the utility grid in HT mode Electricty sold to the utility grid in FL mode

1600 1500 1400 1300 1200 1100 1000 900 800 700 600 500 400 300 200 100 0 1

Heat loss in ET mode Heat loss in FL mode

Heat loss, GJ/a

3 4 5 6 7 8 9 Nominal electric power of the CHP module

10

11

Figure 4. On-site electricity production and heat loss in particular variants of the plant configuration
0,95
Total efficiency of the CHP module

0,90

Net total efficiency of the plant

ET mode HT mode FL mode

0,95

0,90

ET mode HT mode FL mode

0,85

0,85

0,80

0,80

0,75 20 30 40 50 60 70 80 90 100 Nominal electric power of the CHP module 110 120

0,75 20 30 40 50 60 70 80 90 100 Nominal electric power of the CHP module 110 120

Figure 5. Total efficiency (EUF) of the cogeneration module and the whole plant

In the HT and FL modes of CHP module operation the electricity can be sold to the mains. If so, the selling price of surplus electricity will be 140 PLN/MWh. Figure 11 shows the results of the economic analysis in relation to the reference case. It was found that the best mode of operation of the CHP module is electricity tracking. It was also found that there is an optimal solution that consists in installation of CHP module in 80 90 kW electric power range and three gas boilers. The difference in NPV comparing to base case analysis is at the level of investment cost.
450000

Difference in NPV in comparison to reference case, PLN

400000 350000 300000 250000 200000 150000 100000 50000 0

Mode: Electricity tracking Mode: Heat tracking Mode: Full power

20

30

40

50

60

70

80

90

100

110

120

Electric power of the CHP module, kW

Figure 6. Results of the optimization procedure

CONCLUSIONS

The procedure of optimal sizing of a cogeneration plants on the example of demonstration project has been presented in the paper. It has been demonstrated that the optimal solution exists and it can be initially identified by using a general model of the plant and statistical parameters of machinery characteristics. The presented model is also suitable for selecting the optimal mode of the cogeneration module operation. Although the procedure is not a complicated one and quite effective, it must be emphasized that it is only the first step of the analysis. In the next stage technical and economic data for specific engines and boilers have to taken from vendors. General characteristics have to be replaced by the real ones and the whole calculation has to be done again. In some cases the technical and economic real effects can be even more attractive than that one obtained from pre-feasibility study. For the analyzed case the initial study has proved that that the cogeneration makes better solution than the boiler plant only.

NOMENCLATURE

CFt Cash Flow in year t, N electric power or lifetime of the project & heat flux Q & P mass flow of pollutant G & mass flow of water G W r Sn KE Ken Kop J0 Pd L discounted cash flow rate incomes costs of exploitation cost of energy; operational costs investment capital income tax, salvage value

KO&M operation and maintenance cost & Wd stream chemical energy of fuel P & momentary heat shortage; Q LHV lower heating values of fuels economic loses resulted from SR heat shortages labour cost; Kp Kenv environmental cost. kfCHP, kfk cost of fuels; kel() cost of electricity, variable in time according to tariffs. unitary emission fees, kPi unitary cost of water kW

Greek symbols E energy production efficiency binary variable index of allowable heat shortage,

Subscripts CHP related CHP module K related to boiler D related to demand site G related to utility grid 0,1 P plant self use of electricty time. S sale an annual time of the operation str losses min minimal nom nominal i,j,k related to i-th, j-th, k-th module REFERENCES

1. Kalina J. Initial Sizing of the Heat and Electricity Source Under Known Daily Load Profiles Simplified Equation Based Analysis. Proceedings of the Seminar Cogeneration in Industrial and Municipal Energy Systems. Gliwice, Poland, 2003. Skorek J. Analysis of technical and economic effectiveness small-scale cogeneration plants fuelled with gaseous fuels. Silesian University of Technology Publishing, Gliwice 2002. (in Polish) ISBN 83-7335-127-2 2. Witzani M., Pechtl P. Modelling of (cogeneration)-power plants on time dependent power demands of the consumer. Materiay konferencji ASME Cogen-Turbo Conference. Wiede, Austria, August 1995. 3. Yokoyama R., Ito K. Multi-objective Optimization in Unit Sizing of a Gas Turbine Cogeneration Plant. Journal of Engineering for Gas Turbines an Power. Vol. 117. Stycze 1995. This work has been supported by the Polish Committee for Scientific Research under the research grant No. 4 T10B 02225